On the rate of convergence of the ECME algorithm for multiple regression models with t-distributed errors.

KOWALSKI, JEANNE; TU, XIN M.; DAY, ROGER S.; MENDOZA-BLANCO, JOSÉ

Although much work has been done on comparing and contrasting the EM and ECME algorithms, in terms of their rates of convergence, it is not clear what mechanism underlies each and, furthermore, what factors may determine and influence their rates of convergence. In this paper, we examine the convergence rates and properties of these two popular optimisation algorithms as used in computing the maximum likelihood estimates from regression models with t-distributed errors. By approaching this computing problem through the use of two data augmentation schemes, as well as variations of these well-known algorithms, we offer a more composite view on the performance of each.

MAXIMUM likelihood statistics; NEWTON-Raphson method; STOCHASTIC convergence; MULTIPLE regression analysis; DISTRIBUTION (Probability theory); ALGORITHMS

Biometrika, 1997, Vol 84, Issue 2, p269

0006-3444

Academic Journal

10.1093/biomet/84.2.269